UNDERSTANDING MONDRAGON GLOBALIZATION PROCESS: LOCAL … · 2012-06-18 · -Analyse l’impacte...
Transcript of UNDERSTANDING MONDRAGON GLOBALIZATION PROCESS: LOCAL … · 2012-06-18 · -Analyse l’impacte...
UNDERSTANDING MONDRAGON GLOBALIZATION PROCESS: LOCAL JOB
CREATION THROUGH MULTI-LOCALIZATION
Facing globalization threats to community stability
Jose Mari Luzarraga Monasterio Mondragon Unibertsitatea
Dr. Dionisio Aranzadi Telleria Universidad de Deusto [email protected]
Dr. Iñazio Irizar Etxebarria Mondragon Unibertsitatea
Abstract The stability of local communities is threatened by globalization and the international industrial migration process. This article explores Mondragon’s international multi-localization strategy as an effective strategy to avoid de-localization and defend parent cooperatives employment while creating new jobs in developing countries. At the end of 2006, the Mondragon Cooperative Corporation (here after MCC) located in the North of Spain, had 25 Globalized Cooperatives with 65 production plants abroad employing 14.601 people. Based on the Mondragon cooperatives activity between 1996-2006, this paper:
- Measures the relationship of creating employment abroad and defending employment at home.
- Analyses the impact that having production plants abroad has on the number of members vs. non members’ evolution in the parent cooperative and in the company as a whole.
This research includes analysis from 40 production plants in China, India, Mexico, Brazil and Eastern Europe.
Resumen Las comunidades locales están amenazadas por el proceso migratorio industrial internacional de la Globalización. Este artículo analiza la estrategia de multi-localización internacional de Mondragón para evitar la des-localización y defender el empleo en las cooperativas matrices al tiempo que crea empleo en países en vías de desarrollo. En 2006, Mondragon Corporación Cooperativa (MCC en adelante), situada en el norte de España, tenía 25 Cooperativas Globales con 65 fábricas productivas en el extranjero que empleaban a 14.601 personas.
Basado en la actividad de las cooperativas industriales de Mondragón de 1996-2006, el presente articulo:
- Mide la relación existente entre crear empleo exterior y defender empleo en casa.
- Analiza el impacto de tener plantas de producción en el exterior en la evolución del numero de socios vs. no socios en las cooperativas matriz y en total.
Esta investigación contempla el análisis de 40 empresas participadas en China, India, Méjico, Brasil y Europa del Este.
Resumé Les communautés locales sont menacées par le processus migratoire industriel international de la Globalisation. Cet article analyse la stratégie de multilocalisation Internationale de Mondragon pour éviter la deslocalisation et défendre l’emploi au milieu des matrices coopératives et en même tant créer des emplois dans des pays en voie de développement. En 2006, Mondragon Corporation Coopérative (MCC), située au Nord de l’Espagne, possédait 25 Coopératives Globales avec 65 fabriques productives à l’étranger que employaient 14.601 personnes. Basé sur l’activité des coopératives industrielles de Mondragon de 1996-2006, cet article :
- Mesure la relation existante entre la création d’emplois extérieur et la défense des emplois locaux.
- Analyse l’impacte d’avoir des unités de production à l’étranger dans l’évolution du nombre de travailleurs associés et non associés dans la matrice coopérative ainsi qu’internationalement.
Cette investigation contient l’analyse de 40 unités de production en Chine, Inde, Mexique, Brésil, et en Europe de l’Est.
1. INTRODUCTION
As a result of the worldwide acceleration of economy globalization, we are witnessing
an unprecedented creation of multinational production plants in developing countries,
and a North-South and West-East industrial migration process. In the international Free
Trade framework, firms pursue lower cost and market growth expectations in the so-
called BRIC Countries1 (Brazil, Russia, India and China) to succeed in the global
markets.
1 This concept was presented in “Dreaming with BRIC” Goldman Sachs report in 2003. It predicted that Brazil, Russia, India and China would be giant economic forces in the coming century.
2
At the end of 2006 MCC’s international structure employed 83.601 people worldwide.
Over 14.000 are working abroad in the 65 production plants participated in by the 25
Parent cooperatives, with at least one factory abroad (hereafter refers to as GLOBAL
COOPS) 2.
Mondragon’s Globalization Process has several key aspects to be analysed: the
maintenance of the cooperatives values at home and abroad, the impact of foreign
subsidiaries upon the economic development of their regions abroad, the subsidiaries’
workers participation in management, profits and property, and the subsidiaries’ impact
on the parent cooperatives. This article is focused on the impact of international multi-
localization strategy on parent cooperatives’ local jobs and the evolution of cooperative
employment3 .
Industrial capitalist firms, to remain competitive, may assume parent companies
downsizing or closing as strategic alternatives. Mondragon, as a worker cooperative
committed to their coop member’s community, cannot. International multi-localization
Foreign Direct Investment (hereafter refers to as FDI) seems to be the solution to avoid
de-localization, thus defending their local community from the trigger of globalization
threats, “unemployment”.
This article addresses the following questions: First, do Mondragon’s GLOBAL
COOPS through multi-localization FDI, better defend local community stability
creating more jobs in the parent Cooperatives and in their local community? Second,
does Mondragon FDI policy have an impact on the evolution of Parent Cooperatives’
number of coop members vs. non members’ over time?
2 The so called in literature “multinational cooperative holdings” (Cote 2001, Spear 2001) 3 In this article we use indistinctly cooperative employment as cooperative members jobs and non cooperative employment as non cooperative members jobs.
3
In order to answer these questions, we use a specific database based on MCC’s
Companies’ annual activity reports for 1996–2006.
This paper brings together and contributes to three literature mainstreams. The first
includes the paper by Inbroscio, Williamson and Alperovitz (2003) who present
different place-based ownership models as a better counterforce to globalization’s
threats to local community stability. We address this issue, presenting Mondragon and
their international defensive strategy to maintain local community stability as a worker
cooperative empirical example.
The second mainstream investigates the impact of Mondragon’s multi-localization FDI
policy on cooperative principles and non-cooperative employment evolution (Bakaikoa
et al.2004 or Cheney 1999). Our contribution is that we present new data based on a
quantitative analysis of Mondragon’s cooperative member’s evolution in Parent
Cooperatives.
Finally, we present the case of Mondragon in Spain, adding new data to Economic
Geography research on FDI impact on parent companies employment (Barba Navaretti
and Castellani 2003) and on developing countries’ support through employment
(Karnani 2006)
This article is the result of a four-year research project conducted during 2003-2007 by
Mondragon University4. The overall research, as whole is considered a “Single Holistic
case study” (Yin 2003), combining different research methodologies from Mondragon
cooperatives “archival analysis”, more than 100 interviews with MCC employees
worldwide, an analysis of 40 production plants in China, India, Mexico, Brazil, Eastern
4 The whole research was undertaken with the financial support from the Social Economy Department of the Basque Government, in collaboration with Mondragon University.
4
Europe and Turkey and 9 “holistic multiple case studies” of Mondragon international
subsidiaries best practices.
2. THE MAINTENANCE OF THE COOPERATIVE IDENTITY IN THE
GLOBALIZATION PROCESS
Globalization breaks industrial firm’s national and local structure. To remain
competitive in global markets, companies look for new international placements to
reduce costs by investing in lower labour cost countries (LDC) based on an industrial
activity Vertical Integration (VFDI), access to market share in new emerging countries
through Horizontal integration (HFDI) or a combination of both.
This globalization race results in de-localization, companies closing their parent
companies’ industrial activity to immediately open the same activity abroad (Irizar
2006), a re-location when the activity closed does not come from the parent company,
or an expansionist multi-localization, when the new factory opened abroad does not
necessarily mean the closings of an existing factory.5
These industrial localization movements have a substantial impact on local
communities’ stability. On the one hand, foreign firms’ employment and FDI
contribute to socio-economic development and a step forward in terms of poverty
reduction (Karnani 2005, Sachs 2005 or Mandle 2001), although this assumption has
been questioned by numerous scholars6. On the other hand the global market
competition results in manufacturing firms downsizing and shutting down operations in
5 Literature is using “de-localization”, “delocation” or “relocation” being its French translation “délocalisation” Spanish translation “deslocalización” (see IRIZAR 2006, Husson 2005) 6 See Anderson and Cavanagh, Field Guide to the Global Economy, 47 or Dani Rodrik, The new Global Economy and Developing Countries: Making Openness Work, Washington: John Hopkins University Press. Collected also in Williamson, Inbroscio and Alperovitz 2001, p. 46-49
5
those countries with higher labour costs, which has an important social and economical
impact. (Moore 1996).
Recent studies conclude that even though parent companies employment declines, it
would have declined even more if these firms had not invested abroad. (Barba Navaretti
2003). According to the European Commission, “de-localization” is the biggest worry
of the European (E15) community. Economic Geography emerges as an alternative to
this phenomenon, attempting to reinforce the importance of place-based socio-economic
development and the role of community enterprises or industrial worker cooperatives.
(Hudson 2001, Williamson et al 2002).
Here we find, probably, the biggest difference between firms committed or not to its
local community. An industrial capitalist firm, to remain competitive can face the
challenge of globalization with at least two options or strategies: multi-localization or
de-localization. A worker cooperative committed to their coop members community,
has only one option, multi-localization, what is more, it seems to be the solution to
avoid de-localization, defending their local community from the impact of
globalization’s trigger threat, “unemployment”.
But the globalization process is also affecting community-based, democratic and social
enterprise structures. The traditional local structure of worker cooperatives’ was not
adequately prepared to compete in a capitalistic global market. The new global
structures were needed by the cooperatives, causing a “demutualization” or a
“cooperatives hybridization process” (Spear 2001, Borzaga 2001). This has lead to the
appearance of new cooperative realities “Cooperative holding” (Cote 2001, Chaves
6
1999) and new concepts such as “neo-cooperativism” (Larrañaga 2005) or
“coopitalism” (Defourny, 1999).
That was the case for Mondragon, which, as a network of worker cooperatives
committed to the Basque region7 choose to handle foreign investment in a multi-
localization strategy, opting to use FDI on private companies while creating new green-
field affiliate firms abroad or while acquiring partially (through joint ventures) or firms
owned exclusively by a Coop. This fact has been pointed out as a divergence from their
traditional cooperative approach and a mutation of its model in the sense that MCC
behaves as “a traditional capitalist employer operating in low-wage countries” (Huet
1997) with a capitalist expansionism growth illogic for worker-owned business finite
growth ethic, “sacrificing the long-valued buffer zone between them and the turbulence
of the international market” (Cheney 19998).
At the same time, some authors keep a close look on the Mondragon Cooperative
Experience’s evolution, with the hope that “infant” affiliate companies evolve through
adolescence to maturity, following their Parent Companies’ spirit (Vanek 2007), the
possibility of MCC becoming a new “Democratic Multinational Enterprise” (Errasti et
al 2003), or one of the best examples of democracy in the business world (Malone
2001).
At the end of 2006 the 12 Industrial divisions of MCC employed 42.444 people, 19.079
cooperative members,14.601 employees working abroad in 65 production factories of
25 “Multinational Cooperative Companies” (GLOBALCOOPS) and international sales
7 According to MCC mission statement approved on the VII Congress on May 1999.
7
represented 56,7% of their total sales.9 This makes MCC the 7th biggest corporation in
Spain in total sales and the first of the European ranking in “Employee Shareownership”
companies.10
3. MULTI-LOCALIZATION VS. DE-LOCALIZATION
Contrary to what one might think, MCC’s first international steps had no strategic plan
nor did a debate about globalization of cooperatives take place. According to MCC’s
international department director “If we could have maintained our business here we
would not have moved abroad.” and followed with “We were doing what we had to do;
the worst service we could do to the society was to disappear”11. The decisions were
taken to defend local employment in the coming future.
That was the case with one of the first factories created abroad back in 1989. COPRECI
S.Coop12, a cooperative that makes components for household appliances (founded in
1963) was asked to open a factory in Mexico by their US customer in order to continue
being a strategic global provider. Curiously the same US client recently (2006) has
relocated their factory to China, COPRECI Group still provides the same product but
now from their production plant in China. COPRECI Mexico is still open, with a 600
employee’s workforce and lead by a Mexican general manager that recently has become
a “Coop collaborator -member”13.
Once MCC opted to remain competitive in its markets competing at home and abroad
against multinationals (Clamp 2003) the necessity of international multi-localization
became apparent. As far as the issue of competitiveness is concerned, several studies 9 Based on MCC 2006 annual report and its cooperatives annual reports. 10 According to Actualidad Economica No1258, June 2006 and European federation fo Share Ownership 11 Herrasti J., 2005, Mondragon International Congress, Openning Session 12 Copreci S.Coop at the end of 2006 Copreci S.Coop in Aretxabaleta (Mondragon region) had 863 employees being 688 coop members. Copreci Group had 2.030 employees spread worldwide in their 6 factories. As a Copreci spin-off was created in 1973 Orkli S.Coop that at the end of 2006 had 697 employees being 476 coop members. 13 Copreci’s international “collaborator coop members” policy is one of the best practices analyzed in the current research that will be published in future papers.
8
have shown that multinationals perform better than companies limited to one country in
global and local markets.14 As we mentioned before, for a traditional capitalist firm de-
localization or relocation is an alternative, for a worker cooperative is not.
This defensive strategy, which may anticipate a shorter or longer future, has been the
common cause behind all MCC international projects abroad.
Given the fact that the “international dimension” was identified as a fundamental tool
for Mondragon industrial cooperatives competitiveness, a specific Internationalisation
department was created in the new MCC central services structure in 1990. The first
Internationalisation four-year plan was launched for 1991-199415 with one objective, “to
promote the cooperatives’ internationalisation process”.
How has Mondragon been able to maintain employment in the parent companies in the
Mondragon region where labour costs are 5-8 times higher cost than in Eastern Europe
or 20-25 than in China or India?16
The phenomenon of downsizing, plant closures and multiple layoffs has reached crisis
proportions in high cost labour countries in Northern Europe and in the North of the
US17. In order to avoid this same fate, MCC needed to globalize and quickly. A new
slogan in the minds of MCC managers typified their approach: “How many new jobs do
we need to create abroad to maintain one job at home?”18
It has been already almost 20 years since Mondragon’s first production plant was
created abroad and since 1996 there exists homogeneous, comparable data for the 14 See for instance, Doms and Jensen (1998) provide evidence on the US, Criscuolo and Martin (2003) and Girma, Kneller and Pisu (2003) on the UK, Castellani and Zanfei (2003) on Italy and Bellman and Jungnickel (2002) on Germany. 15 MCC Corporative Internationalization Strategic Plan 1999-2003 16 See Eurostat 2004, Irizar 2006 or CH-ina Survey 2005 Behind the China Kaleidoscope 17 See Williamson, Imbroscio and Alperovitz “Making a place for a community” to have a deep theoretical and empirical analysis of US impact of freer international trade. (p 9-16, 31-41) 18 This question was mentioned in almost all the 100 interviews conducted with Mondragon managers
9
Mondragon Cooperatives’ activity. We believe this is sufficient time to do a quantitative
analysis of the social and economical efficiency of this international strategy. Therefore,
we have chosen to analyze MCC Global activity during 1996-2006 for MCC Industrial
Division Cooperatives. Furthermore, we focus on what Ormaechea (co-founder of
MCC19) considers worker’s cooperatives’ most important action, “employment
growth”. This is done at three levels: the level of the Parent cooperative, the level of the
Mondragon Region local community and the global level (Parent Coop + traditional
environment + abroad). We compare those cooperatives with at least a production plant
abroad (GLOBAL COOPS) with the rest of the industrial LOCAL COOPS. At the end
of 2006, we identified 25 GLOBALCOOPS with 65 “production plants abroad”. (See
Figure 1)
Previous research on international multi-localisation impact on employment
As we mentioned before, several studies have analyzed the impact of FDI and
international multi-localization dimension on firms’ competitiveness and economic
performance. FDI impact on employment has not been so widely studied and the debate
is still ambiguous depending on the methodology used and the countries analyzed.
Some of the relevant descriptive studies done in the 90s in the US (Scott 2000, Kletzer
1998 or Sachs and Shatz 1994) conclude that trade has been a net destroyer of domestic
jobs, estimated at 40% of manufacturing jobs lost in the United States during 1978 and
1990. They also conclude that a direct relationship exists between job displacement
rates and import penetration.
19 See IRIZAR, I.,2006, Chapter 4, pp.91-116 where Ormaechea, co-forunder or ULGOR the first cooperative in Mondragon created in 1956 and President of MCC during 1985-90, affirms “Cooperatives have a strong social commitment, it’s most important action, even if it is not always been recognize, their compulsive commitment to maintain and create employment.”
10
But it has only been in the last decade when due to more reliable datasets20 and the use
of more sophisticated regression approaches, that several studies analysing the direct
impact of FDI on parent companies, downsizing and employment destruction have
appeared (Barba Navaretti and Castellani 2003 or Kiyota and Matsuura 2006)).
Barba Navaretti’s studies are especially relevant (Navaretti et al 2003 and 2006). These
studies measure the impact of FDI investment on employment growth using propensity
score matching while analyzing firms in Italy and France during 1999-2002. They also
conclude that the parent companies employment reduction would have been greater if
those firms did not follow a multi-location strategy through FDI. Based on propensity
score matching a counterfactual statement was created to estimate what would be the
hypothetical behaviour in the event that the company would not have invested abroad.
(A version of this methodology is used in the current study. Test 2: H3 and H4)
International multi-localization can respond to a Vertical industrial Integration (VFDI)
or Horizontal industrial integration (HFDI) strategies. Even if differences exist in the
previous approach, according to some studies, both strategies (VFDI and HFDI) tend to
unify. Current FDI investment flows pursue a cost production reduction and a market
access at the same time. That it is the case with continental European FDI into the new
EU members Poland, Czech Republic. (Becker 2005).
Most of the researches try to compare Multinational firms (MNE), LOCAL and those
local companies that assume FDI and for the first time switch to MNEs (hereafter refers
20 The databases used in some of the studies have records since late 90s, that is the case of the French Enquêtes filiales maintained by the Direction of Foreign Economic Relations (DREE) of the French Ministry of Economic and Finances, which has records of firm with more than 20 employees investing abroad for the first time between 1995 and 2000. The Italian Reprint directory, maintained by the Polytechnic of Milan and the Italian Institute for External 1993 and 1998 or the Japanese Kigyou Katsudou Kihon Chousa Houkokusho prepared annually by the Research and Statistics Department, METI (1994–2002).
11
to as SW). It is widely accepted that MNEs behave better in terms of economic
performance.
Another relevant aspect is firm ownership. Conclusions about their performance on
employment destruction differ, being wider accepted than the theory that FOREIGN
companies have greater employment volatility.
Economic geography proposes place-based ownership models as a strong alternative to
local employment instability. “With ownership and control held in a more collective or
community-oriented fashion, such enterprises tend to anchor or root investment more
securely in communities, providing a counterforce to globalization.” (Inbroscio,
Williamson, Alperovitz, 2003.) Until now, this ownership dimension has not been taken
into account in the research conducted on employment growth and multi-location
strategy.
Another aspects such as the use of net job growth or gross job growth21, the influence of
the difference in wages and geographical distance in between parent companies and
their subsidiaries, firm size or the industry and period have been analyzed
Empirical analysis and Methodology
Our analysis of Mondragon’s international multi-localization strategies’ impact on
employment and net job growth addresses the following question: are Mondragon
GLOBAL COOPS through FDI, able to defend better than Mondragon LOCAL COOPS
local community stability by creating more jobs in the parent Cooperatives and in their
local environment?
The analysis includes the following primary variables as indicators of cooperative
employment: 21 Net Job Growth, is defined as the job creation minus job destruction, this is widely accept but might cause misunderstanding, we can have a negative net job growth with a high job creation. Gross job growth takes into account both job creation and job destruction.
12
VARIABLE NAME VARIABLE DESCRIPTION
PT Total number of jobsPIN Number of jobs in Spain
PCOOP Number of jobs in Parent CooperativePEX Number of jobs abroadMEM Number of Coop members
MEMdPT Coop members divided by Total jobs (MEM/PT)MEMdPCOOP Coop members divided by Parent Coop jobs (MEM/PCOOP)
Employment
Coop members vs. Non members
Table 1: Variable description
The sample has been constructed from MCC annual reports. This database includes
cooperative activity reports during 1996-2006 for Mondragon industrial Cooperatives.
For the first time a merged database of Mondragon GLOBAL COOPS exists, allowing
us to compare their performance with other cooperatives.
1999 2002 2006GLO_PT 13,318 21,049 32,041 18,723GLO_PIN 11,884 15,255 18,123 6,239
GLO_PCOOP 11,367 13,832 14,554 3,187
GROWTH 99-06
WORKFORCE AT 31 DEC.
Table 2: MCC Industrial Division workforce: GLOBAL COOPS Source: ex novo, adapted from MCC annual cooperatives reports
Two different tests are conducted in order to measure FDI impact on employment
generation:
Test 1: Comparing GLOBAL COOPS vs. LOCAL COOPS
The working hypotheses are:
H1: “Those Cooperatives with at least one production plant abroad create
more jobs IN THEIR PARENT COOPERATIVE”.
H2: “Those Cooperatives with at least one production plant abroad create
more jobs IN THEIR LOCAL COMMUNITY”.
13
Based on PCOOP and PIN two new variables were defined (PCOOP_99) and (PIN_99).
These two variables were used to confront both groups as a paired sample in two
alternative hypotheses about the distribution: parametric and non parametric
The statistical analysis is based on “paired samples” annually for the period 1999-2006.
Following both a Parametric, “Normal Bivariant Distribution” or a “Non-parametric”
one, comparing both samples’ “means” in order to measure if a “statistically significant
difference” exists or not and the direction of the relationship.
The “parametric” distribution uses a T-student test being K=n-1. For the “non
parametric” distribution we will use the Wilcoxon test.
Test 2: Measuring FDI’s impact on employment growth for SW-GLOBALCOOPS
It is a descriptive approach to Barba Navaretti and Castellani’s (2003) FDI analysis.
The working hypotheses are:
H3: “FDI generates a bigger net-job growth IN THEIR PARENT
COOPERATIVE during the three years after starting FDI”.
H4: “FDI generates a bigger net-job growth IN THEIR LOCAL COMMUNITY
during the three years after starting FDI”.
In order to measure the specific impact of FDI on cooperative employment, we have to
select SW-GLOBALCOOPS. Therefore, we identified those GLOBALCOOPS at the
end of 2006 that were not GLOBAL COOPS in 1999. It can be said that those COOPs
which assumed their first FDI (opened their first production plant abroad) during the
period analyzed (1999-2006) switched from being LOCAL to GLOBALCOOPS.
14
FAGOR INDUSTRIALFAGOR AUTOMATIONMAIEREIKADIKAR WINGROUPLKSFAGOR EDERLANMATZ ERREKACIKAUTXODANOBATSORALUCEORKLI 2004TAJOCOINALDE
2005
1999
2000
2001
2002
YEAR FIRST FDISW-GLOBALCOOPS
Table 3: MCC Industrial Division SW-GLOBAL COOPS (1999-2006). Source: ex novo, adapted from MCC annual reports
We have to compare the performance of the SW firms before exante and after expost
(the FDI). According to Barba Navaretti and Castellani 200322, the expost SW
behaviour needs to be compared with the performance that the company would had if it
did not assume the FDI investment. Firms’ behaviour cannot be replicated as in a
laboratory test , therefore propensity score matching is used in order to construct a
counterfactual firm with which to compare. It is said, based on the company’s previous
performance (sales, employment, etc) the closest LOCAL company in the database is
selected comparing the expost performance of the SW company with this counterfactual
LOCAL company.
Inspired by this approach and taking into account that K=14 for our SW-GLOBAL
COOPS we create a specific counterfactual LOCAL enterprise. The counterfactual
22 This methodological approach was built in 2003 analyzing Italian firms during 1999-2002. Afterwards has been replicate in other countries: 11 European countries (Navaretti and Venables 2004) and Italian and French SWs firms (Navaretti, Castellani and Disdier 2006) with a reliable and robust performance.
15
criterion selected is the average of MCC LOCAL COOPS employment growth during
the specific years for each SW-GLOBALCOOPS investment date.23
Concretely we make an adaptation of (SW) and difference-in-difference estimator
(DID). The DID estimator compares the difference between pre and post-investment
performance growth in both groups. It measures the difference in the change of the
steepness of the performance trajectories for the two groups of firms.
In our case we only include employment performance (L), not including in the analysis
other economic performance (sales, output, etc).
We define (SWI) for the expost comparison in between (SW_GLOBALCOOPS) and
(HIPOTHETIC) behaviour.
We define (LDID) for the comparison in between employment net-job growth expost
and exante for both variables: (SW_GLOBALCOOPS) and (HYPOTHETIC)
Formally, are given by:
α∧
SWI= [∆ 1
3+itL − ∆ 03+itL ]
α∧
LDID= [∆ 1
3+itL − ∆ 13−itL ] − [∆ 0
3+itL − ∆ 03−itL ]
Being:
∆ 13+itL = The mean employment growth achieved by the company on the third year after
the company FDI
∆ 03+itL = The mean employment growth achieved by the counterfactual company on the
third year after the company FDI
23 Please note that we do not apply computer propensity score matching. In future papers will be used integrating Basque Country industrial companies database or Spanish worker cooperatives database.
16
∆1
3−ity = The mean employment growth achieved by the company on the previous three
years before the FDI
∆0
3−ity = The mean employment growth achieved by the counterfactual company on the
previous three years before the FDI
∆⎯Li t = Employment growth achieved in the three years prior to the FDI
i = the specific company K=14
t = the FDI year, being different for each company
The statistical analysis is based on “paired samples” for every SW-GLOBALCOOP.
Following both a Parametric, “Normal Bivariant Distribution” or a “Non-parametric”
one, comparing both samples’ “means” in order to measure if a “statistically significant
difference” exists or not and the direction of the relationship.
The “parametric” distribution uses a T-student test being K=n-1. For the “non
parametric” distribution we will use the Wilcoxon test.
Results
According to the statistical tests results (See Figure 2a):
The H1 and H2 are ACCEPTED, “Those Cooperatives with at least one production
plant abroad create more jobs IN THEIR PARENT COOPERATIVE (H1) and IN
THEIR LOCAL COMMUNITY (H2)”
17
TOTAL COOPS GLOBAL COOPS LOCAL COOPS
PT 20,531 18,723 1,80893.69% 140.58% 21.04%
PIN 8,233 6,239 1,994
41.26% 52.50% 24.71%
P COOP 4,285 3,187 1,098
22.11% 28.04% 13.71% Table 4: Employment growth from 1999 to 2006, measured in net-jobs creation and net-jobs % growth: TOTAL vs. GLOBAL vs. LOCAL COOPS Source: ex novo, adapted from MCC annual reports
In the second test we specifically analyzed the impact on employment growth prior to
the exante and after expost a local coop is assumed to open an international production
plant abroad.
The H3 is NOT ACCEPTED and H4 is ACCEPTED. (See Figure 3a and 3b). Based on
Mondragon SW_GLOBAL_COOPS annual employment growth during the three years
after their first FDI.
We may affirm: “The production activity abroad through FDI created significantly
more jobs IN THE LOCAL COMMUNITY but, even if it created more, is not
significantly bigger than the growth IN THE PARENT COOPERATIVE”.
The employment growth has been compared with the hypothetical performance of
SW_GLOBAL COOPS would have if did not assume the FDI. This was measured by
the new counterfactual variables (HYPOTHETIC) created to compare with SW growth.
18
REAL SW (FDI) HYPOTHETIC (NOT FDI)
PT 3,201 70154.16% 11.86%
PIN 1,263 69621.38% 11.78%
P COOP 412 399
7.80% 7.56% Table 5: Employment growth since FDI to 3 years after FDI measured in net-jobs created and net-jobs % growth: SW_GLOBALCOOPS vs. “HYPOTHETIC”
From the local community stability, Mondragon GLOBAL COOPS defend their local
community employment from globalization threats better. According to MCC,
GLOBAL COOPS also have better economic results as far as “total sales”,
“international sales”, “turnover” and “Parent Cooperatives total sales”.24
In both cases (SW_GLOBALCOOPS) and (HYPOTHETIC) we observed that their
annual growth is reduced, coherent with previous studies that affirm that state that lately
the employment growth in high cost countries is reduced but less that what would be if
these companies did not assume an FDI investment.
Unlike previous studies in the case of Mondragon there is not a Parent companies’
employment net-growth destruction but instead an employment creation. This
conclusion is coherent with Inbroscio, Williamson and Alperovitz’s (2003) conclusions
that presenting place-based ownership models as a superior counterforce to
globalization’s threats to local community stability. Future studies should specifically
confront Mondragon employment growth with neighbour traditional capitalist firms in
Spain.
The previous analysis may prove that Mondragonians were right in their intuition to use
international multi-localization FDI as a tool to maintain and increase cooperative
24 According to J. Herrasti presentation at “Basque economy observatory” on 2007 February the 28tt.
19
employment at home25, but the strategy of introducing capitalistic structures might have
a direct impact on Cooperatives members’ evolution.
4. COOP MEMBERS vs. NON MEMBERS:
Numerous scholars have recently pointed out the risk of Mondragon’s international
policy, which is, growing their non-cooperative employment under affiliated private
economic structures. The original Mondragon commitment to never employ more than
10% non members was exceed using a variety of contractual relationships.
Furthermore, the fact that they do not have a plan to transform these affiliated
companies into a cooperative has been considered a departure from cooperative
principles (Huet 1997).
Therefore our second analysis is focused on Mondragon GLOBAL COOPS26
cooperative member’s evolution and their representation out of the total workforce and
the Parent cooperatives’ workforce. There are previous qualitative analyses on
Mondragon workers democratic control degeneration process (Kasmir 1996, Cheney
1999). Our analysis does not measure nor include the qualitative impact, but the
quantitative evolution of coop members vs. non members.
Empirical analysis
We use a quantitative criterion measuring Mondragon’s number of coop members over
total workforce (MEMdPT) and over Parent Cooperative workforce (MEMdPCOOP).
Here again, we compare GLOBAL COOPS with LOCAL COOPS.
25 On Mondragon University postgraduate Master in Cooperativism in March 2007 we had the opportunity to share with A. Cancelo, MCC President in between 1996-2004, he was positively surprised by the historical archival analysis of the strategies they had to promote at that time. 26 The term that academia has given to this new structure of worker cooperatives parent companies owing private branches “Multinational Cooperative Holdings” . In the current research we will use the term “GLOBAL COOPS”
20
1999 2002 2006GLO_MEM/PT 67.41% 53.09% 38.25%GLO_MEM/PCOOP 78.97% 80.78% 84.22%LOC_MEM/PT 54.96% 60.62% 65.58%LOC_MEM/PCOOP 58.98% 63.91% 74.90%
COOP MEMBERS vs. NON MEMBERS
Table 6: MCC Industrial Division Coop member’s representation evolution Source: adapted from MCC cooperatives’ annual reports
Methodology and working hypotheses
The working hypotheses are:
H5: “Those cooperatives with at least one affiliated production plant abroad do
not necessarily have a smaller number of members out of total workforce
INTERNATTIONALLY”.
H6: “Those cooperatives with at least one affiliated production plant abroad do
not necessarily have a smaller number of members out of total workforce
IN THE PARENT COOPERATIVE”.
In these two hypotheses we follow the same “paired sample” analysis done previously
on H1 and H2.
Results
According to the previous test the results obtained (See Figure 4):
The H5 is REJECTED; the number of members out of total employees is necessarily
smaller for “GLOBALCOOPS” than “LOCALCOOPS” measured year by year for the
period of 1999-2006
The H6 is ACCEPTED; the number of members out of total Parent Cooperative
workforce is not necessarily smaller for “GLOBALCOOPS” than “LOCALCOOPS”
We may conclude that Mondragon’s international policy:
21
- Does not necessarily affect cooperative employment evolution in Parent
Cooperatives. Additionally we may observe that it is bigger in GLOBAL COOPS
increasing from 79% in 1999 to 84% in 2006 while LOCAL COOPS go from 59%
in 1999 to 75% in 2006. Both of them are increasing annually while LOCAL
COOPS is increasing faster.
- Does affect GLOBAL COOPS cooperative employment evolution internationally.
The GLOBAL COOPS increased from 67% in 1999 to 38% in 2006 while LOCAL
COOPS went from 55% in 1999 to 66% in 2006. Therefore an international workers
participation policy is needed.
Nevertheless, a specific analysis should be conducted to measure the qualitative
dimension of democratic workers participation.
5. DATA AND METHODOLOGY LIMITATIONS
In order to properly interpret the previous results the following “data limitations” have
to be taken into consideration:
- The period analyzed (1999-2006) is short, especially for using a statistical test.
- When considering the expansionist internationalization phase of Mondragon
Industrial Coops, all coops except for 5 GLOBAL COOPS are in their 10 first years
of globalization. This fact is critical while interpreting the results.
- Test 1 is “descriptive” and not “cause-effect”: therefore when interpreting, we
assume Post hoc, ergo propter hoc.
- The criteria used in Test 2, does not include propensity score matching, it should be
replicated using the same test, including a database of other industrial companies
and cooperatives outside Mondragon.
22
- Does not include Information about employment in production plants abroad with
less than 3 Million euros, service or sales representative affiliated companies is not
included, nor those with a share of ownership by the parent cooperatives smaller
than 20%.
- The 25 GLOBAL COOPS included are those which at the end of 2006 have a
productive plant abroad, this has to be taken into account while interpreting other
periods such as 1999-2004 alone.
- The influence of economic growth in the Mondragon region, Spain, Europe and
worldwide during the period analyzed, 1999-2006.
- The conclusions are based on merged information; the different stages, sizes and
strategies followed in the globalization process by those 25 GLOBAL COOPS are
not analyzed. (forth coming publications on the same dataset will include this
analysis)
- The employment data analyzed is based on “net growth”: it does not include people
who left the company or those who retired. This fact is especially important when
interpreting growth in the “number of Coop members”.
- The information about cooperatives employment has been gathered based on MCC
cooperatives annual activity reports and specific information provided by MCC.
Information previous to 1999 did not have enough statistical reliability to be
included.
- Information about ULMA Group employment, which joined MCC in 2001, has been
included since that year in order to be coherent with MCC reports. Hypotheses’ tests
including and not including this data have been done with similar results.
- A 95% probability was used on H1, H2, H5 and H6 and a 90% probability on H3
and H4.
23
- Information of new cooperatives created or cooperatives that closed has not been
specially treated.
6. CONCLUSIONS
The current paper “Multi-localization vs. De-localization” demonstrates quantitatively:
- The interest of “international production multi-localization” strategy to deal with
globalization’s threats to community stability (downsizing and de-localization),
based on MCC activity during 1999-2006.
- Mondragon parent cooperatives with foreign production plants, (GLOBAL COOPS)
have a bigger net job growth than Mondragon Local Coops in parent cooperative
(28% > 14%), in the local community (52%>25%) and in total (141% >21%) during
the period analyzed 1999-2006. (H1 and H2)
- Mondragon Cooperatives’ FDI has a direct positive impact on employment growth
in the local community (21%>12%) and a similar behaviour in the parent
cooperative (7,8% = 7,5%) during the three years after FDI investment. (H3 and H4)
- Mondragon GLOBAL COOPS number of coop members over total workforce
internationally has been strongly affected by their international affiliated companies’
policy. (H5)
- Mondragon GLOBAL COOPS’ number of coop members out of total workforce at
Parent cooperatives has not been significantly affected by their international
affiliated companies’ policy. (H6) Being higher on GLOBALCOOPS (84% >75%)
The previous conclusions are coherent with Place-based ownership models as a better
counterforce to Globalization threats to community stability (Inbroscio, Williamson and
24
Alperovitz 2003). Mondragon Cooperatives not only has maintained but has created
new jobs in their local community employment.
FDI does not necessarily result on parent companies jobs destruction, (Barba Navaretti
and Castellani 2004). In the case of Mondragon Industrial Cooperatives, even exist a
positive relationship and creates new jobs.
Horizontal integration has been the main reason for cooperatives multi-localization
abroad in Brazil, Mexico, India or Turkey. China and some specific cases in Eastern
Europe respond to Horizontal and Vertical integration at the same time.
We may say that jobs abroad, even in LDC countries, are not especially unskilled ones.
The General Manager of the 30% of the production plants in these countries is local,
and as an average only one Director from the parent cooperative is working there, so
local teams drive the production plants.
We conclude, on the one hand globalization presents important threats to local
traditional community stability and cooperatives structure. On the other hand leverages
the capacity of wealth creation through employment growth27 on economic developed
countries (20.531 new jobs since 1999) and on the developing ones (12.298 new jobs
since 1999).
But Mondragon Cooperative Experience is not just employment growth, since its very
beginning was pursuing a “New type” of enterprise, coherent with the Human, Christian
and Work community solidarity ideals (Aranzadi 1976). On future papers we study the
reality of those 65 production plants abroad and their current challenges to evolve from
infancy through adolescence to maturity (Vanek 2007).
27 As mentioned previously this contribution is considered by Ormaechea, co-founder of ULGOR, the first cooperative in Mondragon, the most important contribution of cooperatives to society. Additionally according to Karnani 2005 “Fortune at the bottom of the pyramid: a mirage”, creating opportunities of steady employment at reasonable wages is the best way to eradicate poverty.
25
7. FIGURES
GLOBAL COOPS AFFILIATED COMPANIES COUNTRY ACTIVITYCikautxo CZ CZECH R.Cikautxo SK SLOVAK R.Nova Paranoa BRAZILFundicao Brasileiras BRAZILFagor Ederlan - Slovensko SLOVAK R.Maier UK UKMaier CZ CZECH R.Copreci Mexico MEXICOCopreci CZ CZECH R.Copreci Brasil BRAZILCopreci Systems ITALYCopreci Turkia TURKEYCopreci China CHINAFONDEIK sro CZECH R.CZEika sro CZECH R.Eika Mexico MEXICOEika Polska POLAND
FAGOR ELECTRONICA FAGOR Thailand THAILAND SemiconductorsErreka Mex MEXICOMatz-Erreka-Kovoplast CZECH R.Orkli Ningbo CHINAOrkli Do Brasil BRAZILTreboplast CZECH R.Tabiplast POLAND
EMBEGA Embega Polska POLAND Household electrical componentsQuality Lifts Products UKSprinte EV FRANCEElevadores EV Internacional MEXICOBauma POLANDAlpi ITALYNantont Huarong CHINAOrbea USA USALusorbea Montagem Bicic. PORTUGAL
COINALDE Coinalde Polska POLANDShanghai Wingroup CHINANingbo East CHINAArmortech Industries CHINAIrizar Tianjin CHINAIrizar Magreb MOROCO
Irizar Brasil BRAZILIrizar México MEXICOIrizar TVS INDIAIrizar Southern Africa PTY SOUTHAFRICAExtra Electro Menagers MOROCOFagor Mastercook POLANDFagor Brandt FRANCEFagor Brandt Verolanouva ITALYFagor Shanghai CHINAFagor Industrial Mexico MEXICOFagor Industrial Turkey TURKEYFagor Industrial Poland POLANDDanube Internacional FRANCESetricefradue ITALY
MATRICI-BATZ MB Lusitana de Matrices PORTUGAL Injection-moulded and pressed productsMondragon Assembly FRANCEMondragon Assembly GERMANYGaindu Mondragon Assembly MEXICO
FAGOR AUTOMATION BJ Equipment CHINA Electric components for Machine toodsDanomar RUMANIAOverbeck Gmbh GERMANYNewall UK UK
SORALUCE Danobat Bimatec GERMANY Machine toolsEGURKO Egurko - Planerko RUMANIA Machine tools
MONDRAGON ASSEMBLY
DANOBAT
DIKAR WINGROUP
IRIZAR
FAGOR ELECTRODOMESTICOS
FAGOR INDUSTRIAL
TAJO
ELECTRA VITORIA
ULMA CONSTRUCCION
ORBEA
COPRECI
EIKA
MATZ ERREKA
ORKLI
CIKAUTXO
FAGOR EDERLAN
MAIER
Rubber products for automotive and household products
Iron & aluminium products for automotive sector
Plastic products for automotive sector
Household electrical components
Components for cookers: electric hot plates, resistances
Plastic products for automotive and household products
Safety components for household gas products
Plastic products for automotive and household products
Lift manufacture, installation & maintainance
Machine tools
Industrial equipment for construction
Bycycles
Indoor fitness products & sports goods
Luxury coaches
Industrial equipment for construction
Household electrical goods: Kitchens, laundries, refrigerators & mini size goods
Industrial Cooking goods for hotels and restaurants
MONDRAGON ASSEMBLY
*Those ones highlighted in grey are the ones visited and analyzed in the current research
Figure 1: Mondragon 25 GLOBAL COOPS with 65 production plants abroad (2006)
26
Paired Samples Test
55.88815 43.69599 15.44887 19.35738 92.41891 3.618 7 .009
15.70868 13.49216 4.77020 4.42896 26.98841 3.293 7 .013
8.44135 9.65319 3.41292 .37108 16.51163 2.473 7 .043
GLO_PT_99 -LOC_PT_99
Pair1
GLO_PIN_99 -LOC_PIN_99
Pair2
GLO_PCOOP_99 -LOC_PCOOP_99
Pair3
Mean Std. DeviationStd. Error
Mean Lower Upper
95% ConfidenceInterval of the
Difference
Paired Differences
t df Sig. (2-tailed)
Test Statisticsb
-2.366a -2.366a -1.859a
.018 .018 .063ZAsymp. Sig. (2-tailed)
LOC_PT_99 -GLO_PT_99
LOC_PIN_99- GLO_PIN_
99
LOC_PCOOP_99 -
GLO_PCOOP_99
Based on positive ranks.a.
Wilcoxon Signed Ranks Testb.
Figure 2a: Hypotheses H1 and H2 parametric and non parametric tests’ results
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
1999 2000 2001 2002 2003 2004 2005 2006 YEAR
NEW
PIN
JO
BS
CR
EATE
D
TOTAL COOPS
GLOBAL COOPS
LOCAL COOPS
Figure 2b: MCC Industrial D: Local community net job growth (PIN)
27
Test Statisticsb
-.941a -1.569a -3.059a
.347 .117 .002ZAsymp. Sig. (2-tailed)
PCOOP_HYP- PCOOP_
REALPIN_HYPO -PIN_REAL
PT_HYPO -PT_REAL
Based on positive ranks.a.
Wilcoxon Signed Ranks Testb.
Paired Samples Test
******** .145150876 ******** ******** ******** 1.106 11 .292
******** .262710899 ******** ******** ******** 1.973 11 .074******** 1.067696035 ******** ******** ******** 2.386 11 .036
PCOOP_REAL -PCOOP_HYP
Pair 1
PIN_REAL - PIN_HYPOPair 2PT_REAL - PT_HYPOPair 3
Mean Std. DeviationStd. Error
Mean Lower Upper
95% ConfidenceInterval of the
Difference
Paired Differences
t df Sig. (2-tailed)
Figure 3a: Hypotheses H3 and H4 parametric and non parametric tests’ results (SWI)
Test Statisticsc
-.863a -.078b -1.255a
.388 .937 .209ZAsymp. Sig. (2-tailed)
LDID_PCOOP_HYP
- LDID_PCOOP_
REAL
LDID_PIN_HYPO - LDID_
PIN_REAL
LDID_PT_HYPO - LDID_
PT_REAL
Based on positive ranks.a.
Based on negative ranks.b.
Wilcoxon Signed Ranks Testc.
Paired Samples Test
******** .268683975 ******** ******** ******** .485 11 .638
******** .545302292 ******** ******** ******** -.440 11 .668
******** 1.364833094 ******** ******** ******** 1.447 11 .176
LDID_PCOOP_REAL- LDID_PCOOP_HYP
Pair1
LDID_PIN_REAL -LDID_PIN_HYPO
Pair2
LDID_PT_REAL -LDID_PT_HYPO
Pair3
Mean Std. DeviationStd. Error
Mean Lower Upper
95% ConfidenceInterval of the
Difference
Paired Differences
t df Sig. (2-tailed)
Figure 3b: Hypotheses H3 and H4 parametric and non parametric tests’ results (LDID)
28
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
-2 -1 FDI +1 +2 +3 YEAR
EMPL
OYM
ENT:
PO
STS
SW_PCOOP
SW_PT
SW_PIN
HYP_PT
HYP_PCOOP
HYP_PIN
Figure 3c: Merged SW_GLOBALCOOPS vs. HIPOTHETIC employment growth
Paired Samples Test
.01621 .18573 .05151 -.09602 .12844 .315 12 .758
-35.41070 25.19558 8.90798 -56.47474 -14.34667 -3.975 7 .005
.16988 .07882 .02186 .12225 .21751 7.771 12 .000
-10.33647 13.78566 4.87397 -21.86157 1.18863 -2.121 7 .072
GLO_MEMdPT - LOC_MEMdPT
Pair1
GLO_MEMdPT_99 -LOC_MEMdPT_99
Pair2
GLO_MEMdPCOOP -LOC_MEMdPCOOP
Pair3
GLO_MEMdPCOOP_99 -LOC_MEMdPCOOP_99
Pair4
Mean Std. DeviationStd. Error
Mean Lower Upper
95% ConfidenceInterval of the
Difference
Paired Differences
t df Sig. (2-tailed)
Test Statisticsc
-.384a -2.366b -3.180a -1.690b
.701 .018 .001 .091ZAsymp. Sig. (2-tailed)
LOC_MEMdPT -
GLO_MEMdPT
LOC_MEMdPT_99 -
GLO_MEMdPT_99
LOC_MEMdPCOOP
- GLO_MEMdPCOOP
LOC_MEMd
PCOOP_99 -GLO_MEMd
PCOOP_99
Based on positive ranks.a.
Based on negative ranks.b.
Wilcoxon Signed Ranks Testc.
Figure 4: Hypotheses H4 and H5 parametric and non parametric tests’ results
29
8. REFERENCES ARANZADI D., 1976, Cooperativismo industrial como sistema, empresa y experiencia. Universidad de Deusto, Bilbao BAKAIKOA B., ERRASTI A. M. and BEGIRISTAIN A., 2004, “Gobierno y Democracia en los Grupos Empresariales Cooperativos ante la Globalización: El Caso de Mondragon Corporación Cooperativa,” Revista de la economía pública, social y cooperativa, No 048, pp. 53-77, CIRIEC, Valencia. BARBA NAVARETTI G. and CASTELLANI D., 2003, “Investment abroad and performance at home. Evidence from Italian multinationals,” CEPR Discussion Paper N. 4284. BARBA NAVARETTI G., and VENABLES. A.J., 2004, Multinational firms in the world economy, Princeton University Press. BARBA NAVARETTI G., CASTELLANI D., and FISHIER A., 2006, “How does Investing in Cheap Labour Countries Affect Performance at Home? France and Italy,” CERP Discussion Paper, N. 5765. BECKER S.O., EKHOLM K., JACKLE R. MUNDLER M., 2005, Location choice and employment decisions: a comparison of German and Swedish multinationals, Deutsche Bundesbank Discussion Paper, No 08/2005. BELLMAN L. and JUNGNICKEL R. 2002, "(Why) do foreign-owned firms in Germany achieve above-average productivity?, in R. Jungnickel (ed.) Foreign-owned firms: are they different?, Palgrave-Macmillan, Houndmills. BORZAGA C, DEFOURNY J., and others 2004, (ed) The emergence of Social Enterprise, EMES research network, Routledge. CHAVES R., 1999, "Grupos empresariales de la economia social: un análisis desde la perspectiva española", in BAREA, J. et al. Grupos Empresariales de la Economía Social, CIRIEC, Valencia. CHENEY G., 1999, Values at Work: Employee Participation meets Market Pressure at Mondragon, NY: Cornell University Press. CLAMP C., 2003., The Evolution of Management in the Mondragon Cooperatives, Mapping Cooperative Studies in the New Millennium, University of Victoria. COTE D., 2003, Les Holdings coopératives, Evolutions ou Transformation Définitive? Jalons, De Boeck, Paris. DEFOURNY J., 1999, L´Économie Social au Nord et au Sud, Jalons, De Boeck, Paris. DOMS M., JENSEN B., 1998, "Comparing Wages, Skills, and Productivity between Domestically and Foreign-Owned Manufacturing Establishments in the Unites States", in Baldwin R., Lipsey R. and Richardson J. Ed. Geography and Ownership as Bases for Economic Accounting, Studies in Income and Wealth, Vol. 59, The University of Chicago Press, Chicago and London.. ERRASTI A.M., HERAS I., BAKAIKOA B. and ELGOIBAR P., 2003, "The internationalization of cooperatives: the case of the Mondragon Cooperative Corporation", Annals of Public and Cooperative Economics 74:4, pp. 553-584, CIRIEC, Valencia.
30
FIRESTONE D., 2001, "A chief Exporter, and Not At All Pleased About It," The New Yok Times, February 23. HUDSON R., 2000, Production, Place and Environment - Changing Perspectives in Economic Geography, Prentice Hall, London. HUET T., 1997, "Can coops go global? Mondragon is trying", Dollar &Sense, November 1997. IRIZAR, I., 2006, Cooperativas, globalización y deslocalización, Ed. Mondragon Unibertsitatea, Arrasate. KARNANI A., 2006, Fortune at the Bottom of the Pyramid: A Mirage, Ross School of Business, wp 1035, September 2006, University of Michigan. KASMIR S., 1996, The Mith of Mondragon, State University of New York Press, Albany. KIYOTA K. and MATSUURA T., 2006, “Employment of MNEs in Japan: New Evidence,” RIETI Discussion Paper Series 06-E-014. KLETZER L.G., 1998, "International Trade and Job Displacement in U.S. Manufacturing, 1979-1991," in Susan Collins, ed, Imports, Exports, and the American Worker (Washington DC, Brookings Institution, 1998), 423-72. LARRAÑAGA J. 2005: Dilema del cooperativismo en la Era de la Globalización, Ed. Federación de cooperativas de Euskadi. MALONE T. W., 2004, The Future of Work, Boston, Harvard Business School Press MANDEL J., 2001, “Reforming Globalization,” Challenge, March-April 2001, RODICK D., 1999, The New Global Economy and Developing Countries: Making Openness Work, Washington: John Hopkins University Press. SACHS J. and SHARTZ H. J., 1994, "Trade and Jobs in U.S. Manufacturing," Brookings papers on Economic Activity, 1994, vol. 1, 1-84. SCOTT R.E., 2000, "The facts about Trade and Job Creation," Economic Policy Institute Issue Brief, March 24, 2000. SPEAR R. and BORZAGA, C., 2004, "Trends and challenges for Co-operatives and Social Enterprises in developed and transition countries", Edizioni31, Trento 2004. SPEAR, R., 2001, “Globalisation et stratégies des coopératives,” In Coté, D. ed. Holdings Coopératifs: Mutation ou Dérive?, pp.115-138, De Boeck, Brussels. VANEK J., 2007, "A note on the future and dynamics of economic democracy", Advances in the Economics Analysis of Participatory and Labor-managed Firms, Volume, 10, pp. 297-304. WILLIAMSON T., IMBROSCIO D. and ALPEROVITZ G., 2002, Making a place for a community: local democracy in a global era, Routledge, New York. WILLIAMSON T., IMBROSCIO D. and ALPEROVITZ G., 2003, Local Policy Responses to Globalization: Place-Based Ownership Models of Economic Enterprise, The Policy Studies Journal, Vol. 31, No. 1, 2003. YIN R.K., 1984, Case Study Research: Design and Methods, SAGE, US.
31